Abstract Children in the child welfare system possess entitlements to Medicaid, and are the most expensive to insure for their mental health needs. Despite knowing that these children are expensive to insure, Medicaid agencies have little empirical policy guidance on what child characteristics are associated with expenditures, what the magnitude of these expenditures are over the duration of their stay in the welfare system, and the effects of Medicaid policies that can assure access to services while containing costs. This information deficit has resulted in Medicaid budgets being determined on the basis of expediency rather than appropriateness, and this has affected the ability of Medicaid agencies to reduce mental health inequalities among maltreated children. For the first time in the field of child welfare research, members of this research team are assembling a unique data set that links the respondents of a national survey of children coming into contact with child welfare agencies (National Survey of Child and Adolescent Well-Being [NSCAW]) to their Medicaid claims data (Medicaid Analytic Extract), along with information on Medicaid and mental health policies within their counties of residence (from the Caring for Children in Child Welfare [CCCW]). Such a linkage combines the richness of person-level information contained in survey data - such as need for mental health services identified using standardized psychometric measures - with use and expenditure information contained in claims data. County- level information about policies allows the examination of the effectiveness of Medicaid policies in managing expenditures, while controlling for need. Longitudinal data analysis on this uniquely constructed data set will (1) model relationships between various domains of child need, and child-level mental health service utilization and expenditures among NSCAW children;(2) estimate mental health service utilization and expenditure trajectories over the childhood development cycle (ages 0- 18);and (3) examine effectiveness of Medicaid policies targeted at children in child welfare on mental health service utilization and expenditures, and use and expenditure trajectories. These specific aims are consonant with several of AHRQ's funding priorities, including the Child Health Insurance Research Initiative, the Strategic Plan that focuses on children as a priority population, and AHRQ's emphasis on low-income populations and on the elimination of disparities. This application by a new investigator - a former AHRQ dissertation grantee - brings together a highly experienced team of senior investigators and a state Medicaid policymaker, with considerable collective research and practice expertise in psychiatry, child welfare and mental health services research, health and mental health economics and econometrics, and Medicaid claims data analysis to answer these salient questions for a highly vulnerable population. PUBLIC HEALTH RELEVANCE: The mental health consequences of child maltreatment make children in the child welfare system the nation's principal public mental health problem when it comes to children. The burden of financing such care falls upon Medicaid agencies, which have little information on what child and contextual characteristics are associated with increased service use, and what policies can promote desirable child mental health outcomes. This study uses a uniquely constructed data set to model relationships between child characteristics and Medicaid expenditures, the amount of expenditures throughout the childhood development cycle among these children, and the effects of public policies that can serve Medicaid's access and quality agendas for this highly vulnerable population.